10 photonics 4 life 1|2010 TUTORIAL A report in 1929 [1] employed light to “see” tumors buried in deep tissues and basic ideas from those measurements still survive today in biomedical diffuse optics (NIRS, DOS, or DOT). A key contribution to the field was made by J¨obsis in the late 1970s [2] who observed a spec- tral window in the near-infrared (NIR, ~650 –950 nm) wherein photons could travel deep in tissue, as a result of the relatively reduced absorption of water and hemoglobin (see Figure 1) 1 . While the idea of using light to probe tissue morphology or function is an obvious one, it turns out that beyond a relatively superficial (<1mm) level, it is non-trivial. Therefore, these “tradi- tional” techniques remain qualitative (at best) or are fraught with systematic Turgut Durduran, Ph.D., turgut.durduran@icfo.es Medical Optics Group (ICFO-MEDOPT) ICFO – The Institute of Photonic Sciences, Barcelona, Spain www.icfo.es errors (at worst). In a nutshell, for most tissues of interest, the light propagation is effected by scattering and absorption. Three length scales are important: (1) A rather short “scattering length” (~1 - 100μm) which corresponds to the typical distance traveled by photons before they scatter. (2) A longer “random walk step” which corresponds to the typi- cal distance traveled by photons before their directions randomize (~1mm). A wavelength (λ) dependent reduced scattering coefficient (μ´ s (λ)) denotes the reciprocal of the photon transport mean free path. (3) A wavelength dependent absorption length which cor- responds to the typical distance traveled by a photon before it is absorbed. In the near-infrared range, this absorption length is much longer (~200 mm) than scattering and its reciprocal is denoted by the absorption coefficient (μ a (λ)). Based on this understanding, the most critical advance that allowed the develop- ment of biomedical diffuse optics was the acceptance that light transport over long distances is well approximated as a diffusive process [4]. This has provided the necessary physical model to quantitatively separate tissue scattering from tissue absorption, and to accurately incorporate the influence of boundaries and hetero- geneities. It has effectively paved the way for quantitative measurements. Table 1 summarizes the salient features of diffuse optical monitors. Most com- monly reported quantities are oxy- and deoxy-hemoglobin , water and lipid A Brief Tutorial on Biomedical Diffuse Optics Features Accessible quantities Functional physiology Microvasc oxy-hemoglobin concentration (HbO 2 or HbO 2 ) Non-invasive Microvasc deoxy-hemoglobin concentration (Hb or ΔHb) Fast (ms to sec) Microvasc total hemoglobin concentration (THC or ΔTHC) Safe (no ionizing radiation) Microvasc cerebral blood volume (CBV or ΔCBV) Portable Instrumentation Microvasc blood oxygen saturation (StO 2 ) Suitable for multi-modality use Oxygen extraction fraction (OEF) Relatively inexpensive Microscopic lipid concentration Deep tissue (1–5 cms) Microscopic water concentration Deep: Low resolution (0.5 –1cm) Microscopic scattering (μ` s ) Surface (100μm –1000μm) Blood flow (BF or rBF) Surface: High resolution (μm) Cerebral metabolic rate of O 2 extraction (CMRO 2 or rCMRO 2 ) Oxidized cytochrome c-oxidase Contrast agent concentration Table 1: Salient features of diffuse optical monitors. “Microvasc” denotes “Microvascular”, prefix “r” implies “relative change” and prefix implies “differential change”. 1 As a historical side-note, the spectra of oxy-hemoglobin in the intra-red was reported in the first article of the first volume of Physical Review Series I [3]. Special thanks to I J Bigio who has pointed this out.